In most plants a wide variety of physiological processes are influenced by light acting through one or more photoreceptors. Phytochrome, one of the best characterized photoreceptors, is a chromoprotein that exists in either of two spectrally distinct and photointerconvertible forms: P.sub.r and P.sub.fr. P.sub.fr is the biologically active form of the molecule. P.sub.r is converted to P.sub.fr by red light which then induces physiological responses; far-red light, which drives P.sub.fr back to P.sub.r, will usually negate the inductive effect of red light, if it is given sufficiently soon after the red. Inductions by small amounts of red light and reversibility with far-red light are standard criteria by which phytochrome involvement in a process is assessed (Shropshire and Mohr, (eds.) (1983) Photomorphogenesis, Encyc. Plant Physiol. 16:3-19) .
Many of the physiological responses to light involve changes in gene activity; different RNA sequences are synthesized when dark-grown plants are transferred to light. Responses of a given gene to light may not always be the same in different plants. There is considerable diversity of responses for different genes and for the same gene under different physiological conditions. For example, it has been shown that the mRNA level of the ribulose-1,5-bisphosphate carboxylase, small subunit (rbvcS) gene may be very high in several plant species (peas, tobacco, maize) yet quite low in others (barley, mungbean) (Tobin and Silverthorne (1985) Ann. Rev. Plant. Physiol. 36:569-593). Also, the contribution to total mRNA by different members of a gene family may be variable and may be influenced by light (Kaufman et al. (1985) Plant Physiol. 78:388-393). The molecular basis for such differential responsiveness remains unknown.
It is generally believed that there are specific DNA sequences which can confer light responsiveness on a gene and which can determine the various different patterns of light induction. The most abundant mRNAs in leaves of higher plants encode the chlorophyll a/b binding (Cab) protein and the rbcS. Nuclear genes for these mRNAs are expressed in an organ-specific manner and their expression is induced by light acting through phytochrome.
Cab proteins associate with chlorophylis a and b to form the light harvesting complex which absorbs energy from light and transfers the resultant excitation energy to photosystems I and II (Arntzen (1978) Curr. Top. Bioenerget. 8:111-160). The Cab proteins are synthesized by free cytoplasmic ribosomes as precursor proteins, which contain an amino acid terminal extension or transit peptide. This transit peptide is cleaved during membrane translocation to produce the mature polypeptide (Schmidt et al. (1981) J. Cell. Biol. 91:468-478).
Cab proteins are encoded by small gene families as part of the nuclear genome in several plant species (Dunsmuir (1985) Nucl. Acid Res. 13:2503-2518). Polypeptides of different sizes appear to be encoded by these genes.
To investigate the regulation and expression of Cab genes in more detail, Lamppa et al. (1985) Mol. Cell. Biol. 5:1370-1378, sequenced a genomic clone for a major Cab polypeptide from wheat. This gene encoded a 70 nucleotide 5'-nontranslated spacer, a 34 amino acid transit peptide and a mature coding protein of 232 amino acid residues. The molecular weight of the precursor polypeptide was 28,560 daltons. The wheat Cab gene structure was compared with that of a Cab gene from pea (Cashmore (1984) Proc. Natl. Acad. Sci. 81:2960-2964). The coding regions of genomic clones from pea (dicot) and wheat (monocot) showed 90% homology, whereas the wheat transit peptide was 3 amino acids shorter than that found in pea. Furthermore, within the various Cab genes, sequence divergence was observed in the 5'-nontranslated leader region as well as in the transit peptide.
The schedule of expression for Cab mRNAs and polypeptides during maize leaf development in light and darkness was determined by Nelson et al. (1984) J. Cell. Biol. 98:558-564. In maize, Cab polypeptides increased during development in light from very low or undetectable levels to become one of the most abundant leaf proteins. In this study it was found that the maize Cab protein accumulation was absolutely dependent on light and only low levels (&lt;0.5% of final levels) of Cab mRNA could be detected in dark-grown tissue, but these mRNA levels increased up to 200 fold with illumination. In many other plants the same overall result was obtained--that the expression of Cab gene is regulated by light, with little or none of the pre-Cab mRNA being present in dark-grown tissue (Cashmore (1984) supra).
The maize Cab multigene family is thought to consist of at least 12 members (Sheen and Bogorad (1986) Proc. Natl. Acad. Sci. 83:7811-7815). Six members of the maize Cab multigene family were identified and studied selectively for expression patterns (Sheen and Bogorad (1986) supra). In cells of illuminated dark-grown maize leaves, transcripts of these six Cab genes were present at vastly different levels and accounted for about 95% of total Cab mRNA. After 24 hours of greening, Cab-ml was the most highly expressed gene, contributing approximately 30% of the total Cab mRNA; while Cab-m5, Cab-m2, Cab-m3, Cab-m4, and Cab-m6 contributed 20%, 18%, 15%, 8% and 4%, respectively. Of these six maize Cab genes, Cab-m1, the most highly expressed gene of the family in greening leaves, and Cab-m6, the least expressed, were both strongly induced by light. Cab-m5 was induced by light less strongly, whereas Cab-m2, Cab-m3 and Cab-m4 were induced by light only slightly. mRNAs of two Cab genes (Cab-m4, and to a lesser extent Cab-m3), were present in etiolated seedlings prior to illumination and did not show strong light responsiveness.
The studies of Sheen and Bogorad (1986) supra can be compared to those of Nelson et al. (1984) supra. In showing that accumulation of Cab polypeptides in whole maize leaves was absolutely light-dependent while the mRNA, present at a low level in the dark was dramatically stimulated by light, Nelson et al. (1984) supra may have described the expression patterns of the Cab-m1 and Cab-m6 genes of illuminated dark-grown maize seedlings as reported by Sheen and Bogorad (1986) Supra. The other four Cab genes described by Sheen and Bogorad (1986) Supra did not show such dramatic response to light. These studies of Sheen and Bogorad (1986) supra suggested that the six members of the maize multigene family are not regulated through a common mechanism. Whether the various Cab proteins encoded by the different genes and under different regulatory controls have different functions or whether the proteins are interchangeable is not known.
The regulatory functions of the 5'-flanking sequences of the pea Cab gene were investigated by Simpson et al. (1985) EMBO J. 4:2723-2729. Two fragments containing either 2.5 or 0.4 kbp of the 5' flanking sequences upstream to the transcription start site of the pea Cab gene were fused to the neomycin phosphotransferase II (NPT(II)) gene from Tn5 as an enzymatic reporter. It was shown that 0.4 kbp of the upstream flanking sequences of the pea Cab gene were sufficient for both organ-specific and light-regulated expression of the chimeric constructs in transformed tobacco plants. In addition, it was shown in this study that sequences further upstream (-2.5 to -0.4 kbp from the transcription start site) are required for enhancement (regulation) of the overall level of expression. However, when the 5'-flanking sequences of the Cab gene were replaced with the nopaline synthase (nos) promoter (the nos promoter is known to be constitutively expressed in all tissues of transformed plants but is not normally photoresponsive), the level of NPT(II) activity in the transgenic plants was no longer light responsive. These observations indicated that: (a) DNA sequences important in defining light/dark differences are located in the 5' flanking region, although intragenic or 3' sequences may also have important roles in regulating gene function, and (b) control is at the transcriptional level.
Continuing work on the 5' flanking sequences of the pea Cab gene revealed the presence of an enhancer-like element in the upstream region within 400 bp of the transcription start site (Simpson et al. (1986) Nature 323:551-554). The enhancer activity was found to reside within a 247 bp fragment from -100 to -347 bp in the 5' flanking region of the AB80 gene. This 247 bp sequence was cloned in both orientations upstream of the constitutively expressed nos promoter and fused to the NPT coding sequence. In control constructs devoid of the 247 bp sequence, transgenic plants were unresponsive to light conditions. In contrast, when the 247 bp sequence was inserted in either direction into fusion constructs, the resultant transgenic plants showed four to eight times higher NPT activity under light than under dark growth conditions. These experiments showed that the presence of the pea AB80 upstream element confers light-inducible properties on the recombinant nos/NPT gene and, in addition, showed that insertion of two copies of the upstream element 5' to the nos promoter appeared to have an additive effect.
Other work by Nagy et al. (1986) Phil. Trans. R. Soc. Lond. 314:493-500) with the Cab gene from wheat suggested that an important cis-acting regulatory element was located between -90 and -144 of the Cab-1 gene. These authors also suggested that by analogy with other regulated genes, the Cab upstream segment contained one or more light responsive elements comprising the `GT` sequence motif resembling the SV40 enhancer core sequence (Gluzman (1983) in Enhancers and eukaryotic gene expression, Y. Gluzman (ed.) Cold Spring Harbor, N.Y., pp. 27-32) to which cognate transacting factors would bind.